As a means for connecting information devices to one another, a wired or wireless LAN (Local Area Network) has become widespread and has been widely used in homes and offices.
In such a network, TCP/IP protocol is generally used. Data communication is realized by exchanging data packets complying with this protocol. A terminal device connected to a network is assigned an address called an IP address for identifying the terminal device on the network. A data packet contains the IP addresses of a transmission source and destination. That is, in a network designed to perform communication by using TCP/IP, IP addresses must be set for the respective terminal devices which communicate with one another.
As a method of setting IP addresses for terminal devices, there is available a method of automatically assigning a proper IP address to a terminal device newly connected to a network by using a protocol called DHCP (Dynamic Host Configuration Protocol), in addition to a method of setting fixed IP addresses for terminal devices in advance.
When DHCP is to be used, IP addresses are managed by a DHCP server on a LAN. Upon receiving a request to assign an IP address from a terminal device newly connected to the network, the DHCP server selects an unused IP address from a plurality of stored IP addresses and transmits it to the terminal device. The use of DHCP makes it possible to assign IP addresses to only terminal devices connected to the network without assigning any IP addresses to terminal devices which are not connected to the network. Therefore, an IP address resource can be efficiently used as compared with the case wherein fixed IP addresses are assigned to the respective terminal devices.
When a portable terminal device such as a notebook PC is to be used, the use of DHCP makes it unnecessary to perform complicated operation to change the IP address of a terminal device every time it connects to a different network. This makes it possible for the user to use various networks regardless of the place where the terminal device is used.
In the above network, however, an unauthorized user who is not permitted to connect to the network can use the network by connecting his/her terminal device to the network without permission. This allows unauthorized access to a server, or the like, on the LAN. In order to solve this problem, it is preferable to authenticate the user of a terminal device when the terminal device is connected to the network.
In general, such a system often uses a method of preparing, on the LAN, an authentication server storing authentication information comprising user IDs, password information, and the like, and performing authentication by causing a terminal, which tries to connect to the network, to transmit authentication information to the authentication server. Introducing such an authentication means makes it possible to solve the above problem associated with security on the network.
Assume that a given wireless terminal device whose authentication information is not registered in the authentication server need not access to a server or the like on the LAN but wants to communicate with another wireless terminal device through a packet relay apparatus such as a wireless access point or router by using TCP/IP. In this case, no packet is relayed to the DHCP server because of an authentication failure, and hence any IP address from the DHCP server cannot be assigned to the terminal device. As a consequence, TCP/IP-based communication cannot be performed.
This problem can be avoided by fixing an IP address to the terminal device in advance. This, however, makes operation complicated. In addition, the setting information of a network to which connection is to be made must be known in advance, resulting in an increased work load on the user.